Lowintensity cognitivebehaviour therapy clinical ...centaur.reading.ac.uk/71064/6/journal.pmed.1002337.pdf · part of the peer-review process for funding, peer-reviewers identified

Lowintensity cognitivebehaviour therapy interventions for obsessivecompulsive disorder compared to waiting list for therapistled cognitivebehaviour therapy: 3arm randomised controlled trial of clinical effectiveness Article

Published Version

Creative Commons: Attribution 4.0 (CCBY)

Open Access

Lovell, K., Bower, P., Gellatly, J., Byford, S., Bee, P., McMillan, D., Arundel, C., Gilbody, S., Gega, L., Hardy, G., Reynolds, S., Barkham, M., Mottram, P., Lidbetter, N., Pedley, R., Molle, J., Peckham, E., KnoppHoffer, J., Price, O., Connell, J., Heslin, M., Foley, C., Plummer, F. and Roberts, C. (2017) Lowintensity cognitivebehaviour therapy interventions for obsessivecompulsive disorder compared to waiting list for therapistled cognitivebehaviour therapy: 3arm randomised controlled trial of clinical effectiveness. PLOS Medicine, 14 (6). e1002337. ISSN 15491676 doi: https://doi.org/10.1371/journal.pmed.1002337 Available at http://centaur.reading.ac.uk/71064/

It is advisable to refer to the publisher’s version if you intend to cite from the work. See Guidance on citing .

http://centaur.reading.ac.uk/71187/10/CentAUR%20citing%20guide.pdf

To link to this article DOI: http://dx.doi.org/10.1371/journal.pmed.1002337

Publisher: Public Library of Science

All outputs in CentAUR are protected by Intellectual Property Rights law, including copyright law. Copyright and IPR is retained by the creators or other copyright holders. Terms and conditions for use of this material are defined in the End User Agreement .

www.reading.ac.uk/centaur

CentAUR

Central Archive at the University of Reading

Reading’s research outputs online

http://www.reading.ac.uk/centaur

http://centaur.reading.ac.uk/licence

RESEARCH ARTICLE

Low-intensity cognitive-behaviour therapy

interventions for obsessive-compulsive

disorder compared to waiting list for

therapist-led cognitive-behaviour therapy:

3-arm randomised controlled trial of clinical

effectiveness

Karina Lovell1*, Peter Bower2, Judith Gellatly1, Sarah Byford3, Penny Bee1,

Dean McMillan4, Catherine Arundel5, Simon Gilbody4, Lina Gega4, Gillian Hardy6,

Shirley Reynolds7, Michael Barkham6, Patricia Mottram8, Nicola Lidbetter9,

Rebecca Pedley1, Jo Molle10, Emily Peckham5, Jasmin Knopp-Hoffer2, Owen Price1,

Janice Connell11, Margaret Heslin3, Christopher Foley12, Faye Plummer13,

Christopher Roberts14

1 Division of Nursing, Midwifery and Social Work, School of Health Sciences, Faculty of Biology, Medicine

and Health, The University of Manchester, Manchester Academic Health Science Centre, Manchester, United

Kingdom, 2 NIHR School for Primary Care Research, University of Manchester, Manchester, United

Kingdom, 3 King’s Health Economics, Institute of Psychiatry, Psychology & Neuroscience, King’s College

London, London, United Kingdom, 4 Hull York Medical School & Department of Health Sciences, University

of York, York, United Kingdom, 5 Department of Health Sciences, University of York, York, United Kingdom,

6 Centre for Psychological Services Research, Department of Psychology, University of Sheffield, Sheffield,

United Kingdom, 7 School of Psychology, University of Reading, Reading, United Kingdom, 8 Cheshire &

Wirral Partnership, NHS Foundation Trust, Wallasey, United Kingdom, 9 Anxiety UK, Manchester, United

Kingdom, 10 Norwich Medical School, University of East Anglia, Norwich, United Kingdom, 11 ScHARR,

University of Sheffield, Sheffield, United Kingdom, 12 Department of Public Health & Primary Care,

Cambridge University, Cambridge, United Kingdom, 13 Bradford Institute for Health Research, Bradford,

United Kingdom, 14 Centre for Biostatistics, University of Manchester, Manchester, United Kingdom

* [email protected]

Abstract

Background

Obsessive-compulsive disorder (OCD) is prevalent and without adequate treatment usually

follows a chronic course. “High-intensity” cognitive-behaviour therapy (CBT) from a spe-

cialist therapist is current “best practice.” However, access is difficult because of limited

numbers of therapists and because of the disabling effects of OCD symptoms. There is a

potential role for “low-intensity” interventions as part of a stepped care model. Low-intensity

interventions (written or web-based materials with limited therapist support) can be provided

remotely, which has the potential to increase access. However, current evidence concerning

low-intensity interventions is insufficient. We aimed to determine the clinical effectiveness

of 2 forms of low-intensity CBT prior to high-intensity CBT, in adults meeting the Diagnostic

and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for OCD.

PLOS Medicine | https://doi.org/10.1371/journal.pmed.1002337 June 27, 2017 1 / 19

a1111111111

a1111111111

a1111111111

a1111111111

a1111111111

OPENACCESS

Citation: Lovell K, Bower P, Gellatly J, Byford S,

Bee P, McMillan D, et al. (2017) Low-intensity

cognitive-behaviour therapy interventions for

obsessive-compulsive disorder compared to

waiting list for therapist-led cognitive-behaviour

therapy: 3-arm randomised controlled trial of

clinical effectiveness. PLoS Med 14(6): e1002337.

https://doi.org/10.1371/journal.pmed.1002337

Academic Editor: Phillipa J. Hay, Western Sydney

University, AUSTRALIA

Received: November 11, 2016

Accepted: May 26, 2017

Published: June 27, 2017

Copyright: © 2017 Lovell et al. This is an open

access article distributed under the terms of the

Creative Commons Attribution License, which

permits unrestricted use, distribution, and

reproduction in any medium, provided the original

author and source are credited.

Data Availability Statement: All relevant data are

within the paper and its Supporting information

files.

Funding: KL received funding from the National

Institute for Health Research (NIHR) Health

Technology Assessment (HTA) programme (Ref:

09-81-01 https://www.journalslibrary.nihr.ac.uk/

programmes/hta/098101/#/). KL has full access to

all the data in the study and had final responsibility


http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pmed.1002337&domain=pdf&date_stamp=2017-06-27







http://creativecommons.org/licenses/by/4.0/

https://www.journalslibrary.nihr.ac.uk/programmes/hta/098101/#/


Methods and findings

This study was approved by the National Research Ethics Service Committee North West–

Lancaster (reference number 11/NW/0276). All participants provided informed consent to

take part in the trial. We conducted a 3-arm, multicentre randomised controlled trial in pri-

mary- and secondary-care United Kingdom mental health services. All patients were on a

waiting list for therapist-led CBT (treatment as usual). Four hundred and seventy-three eligi-

ble patients were recruited and randomised. Patients had a median age of 33 years, and

60% were female. The majority were experiencing severe OCD. Patients received 1 of 2

low-intensity interventions: computerised CBT (cCBT; web-based CBT materials and limited

telephone support) through “OCFighter” or guided self-help (written CBT materials with lim-

ited telephone or face-to-face support). Primary comparisons concerned OCD symptoms,

measured using the Yale-Brown Obsessive Compulsive Scale–Observer-Rated (Y-BOCS-

OR) at 3, 6, and 12 months. Secondary outcomes included health-related quality of life,

depression, anxiety, and functioning. At 3 months, guided self-help demonstrated modest

benefits over the waiting list in reducing OCD symptoms (adjusted mean difference = −1.91,

95% CI −3.27 to −0.55). These effects did not reach a prespecified level of “clinically signifi-

cant benefit.” cCBT did not demonstrate significant benefit (adjusted mean difference =

−0.71, 95% CI −2.12 to 0.70). At 12 months, neither guided self-help nor cCBT led to differ-

ences in OCD symptoms. Early access to low-intensity interventions led to significant reduc-

tions in uptake of high-intensity CBT over 12 months; 86% of the patients allocated to the

waiting list for high-intensity CBT started treatment by the end of the trial, compared to 62%

in supported cCBT and 57% in guided self-help. These reductions did not compromise lon-

ger-term patient outcomes. Data suggested small differences in satisfaction at 3 months,

with patients more satisfied with guided self-help than supported cCBT. A significant issue

in the interpretation of the results concerns the level of access to high-intensity CBT before

the primary outcome assessment.

Conclusions

We have demonstrated that providing low-intensity interventions does not lead to clinically

significant benefits but may reduce uptake of therapist-led CBT.

Trial registration

International Standard Randomized Controlled Trial Number (ISRCTN) Registry

ISRCTN73535163.

Author summary

Why was this study done?

• Many people suffer from obsessive-compulsive disorder (OCD), and if they do not get

treatment, it can be a long-term problem.

• Although cognitive-behaviour therapy (CBT) with a therapist is effective, many people

struggle to get access because of limited numbers of therapists.

Low-intensity CBT interventions for obsessive-compulsive disorder: A randomised controlled trial


for the decision to submit for publication. SB was

funded for this research by the NIHR Health

Technology Assessment Programme. The funders

had no role in the study design, data collection and

analysis, the writing of the report, or the decision to

submit the paper for publication.

Competing interests: MB was the lead investigator

in the development of the CORE-OM, a measure

used in the reported study. Funding was from the

Mental Health Foundation during the period 1995–

1998. MB has never received any financial payment

in relation to the measure and holds no financial

interest in it. In January 2016, the CORE-OM was

placed under Creative Commons Agreement. NL is

Chief Executive of Anxiety UK and Director of

Mental Health for the Big Life group and Self Help

(Services). DM is co-applicant on UK National

Institute of Health Research (NIHR) grants, in

addition to the OCTET trial, funded by the NIHR

Health Technology Assessment funding stream. As

part of the peer-review process for funding, peer-

reviewers identified by the NIHR may provide

comments on the design, analysis, and

interpretation of results. However, ultimately all

stages of the research are under the control of the

research team. There is no influence on the writing

of the paper or the decision to submit the results

for publication. DM is additionally a board member

of UK NIHR Research for Patient Benefit (RfPB)

Yorkshire and Humberside Committee. This Board

was not involved in any way in the funding of the

OCTET trial.

Abbreviations: cCBT, computerised cognitive-

behaviour therapy; CBT, cognitive-behaviour

therapy; CONSORT, Consolidated Standards of

Reporting Trials; CORE-OM, Clinical Outcomes in

Routine Evaluation; CSQ-8, Client Satisfaction

Questionnaire; DSM-IV, Diagnostic and Statistical

Manual of Mental Disorders, Fourth Edition; EQ-5D,

EuroQol five dimensions questionnaire; GAD-7,

Generalised Anxiety Disorder 7-item; GSH, guided

self-help; HTA, Health Technology Assessment;

ICC, intracluster correlation; MCS, Mental

Component Summary; MHRN, Mental Health

Research Network; NHS, National Health Service;

OCD, obsessive-compulsive disorder; OCTET,

Obsessive Compulsive Treatment Efficacy Trial;

OR, odds ratio; PCS, Physical Component

Summary; PHQ-9, Patient Health Questionnaire;

PWP, Psychological Wellbeing Practitioner; SF-36,

Short Form-36; Y-BOCS-OR, Yale-Brown

Obsessive Compulsive Scale–Observer-Rated; Y-

BOCS-SR, Yale-Brown Obsessive Compulsive

Scale–Self-Report; WSAS, Work and Social

Adjustment Scale.

http://www.isrctn.com/ISRCTN73535163


• Low-intensity versions of CBT (written or web-based materials with limited therapist

support) may increase access to care, but evidence of their effectiveness is limited.

What did the researchers do and find?

• We tested 2 low-intensity versions of CBT in a trial (guided self-help and supported

computerised cognitive-behaviour therapy [cCBT]), testing their impact on patients

with OCD when provided prior to CBT with a therapist.

• Neither low-intensity version of CBT led to clinically significant benefits in patient

outcomes.

• Access to “low-intensity” interventions led to significant reductions in uptake of CBT

with a therapist over 12 months.

• More patients were satisfied with guided self-help than supported cCBT.

What do these findings mean?

• Providing low-intensity interventions does not lead to clinically significant benefits but

may reduce uptake of therapist-led CBT.

• These findings from a large pragmatic trial showing no clinical benefit from low-inten-

sity treatments are in contrast to other studies published recently.

Introduction

Obsessive-compulsive disorder (OCD) has an estimated lifetime prevalence of 2%–3% and is

rated among the top 10 causes of disability worldwide, with an estimated US$8.4 billion attrib-

utable to OCD in the United States [1]. Providing accessible and effective care for OCD is a

priority worldwide.

However, there is evidence that people with OCD struggle to access treatment, with consistent

reports of a marked delay between OCD onset and management. One study found a 10-year gap

between onset and seeking help and 17 years between onset and receiving effective help [2].

In OCD, both medication and psychological therapy are effective, with the “gold standard”

psychological therapy intervention being therapist-led cognitive-behaviour therapy (CBT) [3],

with 1-hour weekly sessions delivered predominantly face-to-face over 12–16 weeks. However,

it is relatively costly, and the limited availability of specialist therapists means that access can

be poor, with long waiting times. Additionally, OCD is characterised by intrusive, unwanted,

recurrent, and distressing thoughts, images, or impulses (i.e., obsessions) and repetitive actions

or rituals (compulsions). These obsessions and compulsions can make it more difficult for

patients to engage with treatment because of fears of contamination or causing harm to others.

Conventional ways of delivering psychological therapy are being challenged. Health systems

under financial pressure need to manage demand more effectively through new methods of

delivery [4], and innovation is needed to meet the needs of diverse patient populations with

complex needs. Research has shown how conventional therapist-led CBT can be delivered

effectively in low-intensity forms including guided self-help (written CBT materials with lim-

ited telephone or face-to-face support) or computerised CBT (cCBT; web-based CBT materials




and limited telephone support). Both forms are potentially cheaper and more accessible than

conventional therapist-led CBT and demonstrate evidence of effectiveness in a range of disor-

ders [5,6]. Low-intensity CBT interventions for OCD may provide more rapid relief of symp-

toms, reduce the need for more expensive therapist-led CBT, and encourage more efficient use

of healthcare resources when delivered as part of a stepped care system [7].

At the time this study was commissioned, the evidence base for low-intensity interventions

in OCD was far from definitive. Much of the evidence for guided self-help was based on small

open or uncontrolled studies [8,9] or comparisons of nonguided self-help with guided self-

help [10]. A systematic review of cCBT for OCD found only 4 studies [11]. cCBT reduced ritu-

als and obsessions and improved functioning and was more effective than attention control

[12], but not as effective as therapist-led CBT. Clearly, the potential benefits of both guided

self-help and cCBT need to be demonstrated in large-scale trials.

We conducted a randomised trial for patients with OCD, allocating patients to either (a)

guided self-help prior to therapist-led CBT, (b) cCBT prior to therapist-led CBT, or (c) a wait-

ing list for therapist-led CBT only. We aimed to provide a definitive answer to the following

questions:

1. Does access to guided self-help or cCBT provide more rapid improvement in OCD symp-

toms at 3 months compared to a waiting list for therapist-led CBT?

2. Does access to guided self-help or cCBT improve patient satisfaction at 3 months compared

to a waiting list for therapist-led CBT?

3. Does access to guided self-help or cCBT prior to therapist-led CBT provide longer-term

improvement in OCD symptoms at 12 months compared to therapist-led CBT alone?

4. Does access to guided self-help or cCBT reduce uptake of therapist-led CBT over 12

months?

Methods

This study was approved by the National Research Ethics Service Committee North West–

Lancaster (reference number 11/NW/0276). All participants provided informed consent to

take part in the trial.

Study design and participants

We conducted a pragmatic trial, delivered in routine service settings, to provide a balance

between internal and external validity and maximise relevance to clinical guidelines [13,14].

The study protocol has been published [15]. The study is reported as per Consolidated Stan-

dards of Reporting Trials (CONSORT) guidelines, as described in the CONSORT checklist (S2

Text). Potential participants were most frequently identified by administrative and clinical

staff in primary-and secondary-care screening waiting lists, although self-referral was used at 1

site to increase recruitment (via adverts in newspapers, community facilities, and social

media). Potential participants were provided with an information pack. Those who provided

consent to contact took part in a telephone eligibility screen to determine if they were over 18

years old and not currently receiving a psychological therapy for OCD or experiencing severe

and distressing psychotic symptoms. Participants passing the initial screen were offered a face-

to-face eligibility appointment.

We included patients who were (1) aged 18+ years, (2) able to read English, (3) currently

waiting for access to therapist-led CBT, (4) meeting DSM-IV criteria for OCD (assessed using




the Mini-International Neuropsychiatric Interview [16]), and (5) scoring 16+ on the Yale-

Brown Obsessive Compulsive Checklist–Self-Report (Y-BOCS-SR [17]).

We excluded patients (1) experiencing active suicidal or psychotic thoughts, (2) meeting

DSM-IV alcohol or substance dependence criteria, (3) receiving psychological treatment for

OCD, or (4) with language difficulties that would preclude participation.

Randomisation and masking

Patients were randomised (ratio 1:1:1) via a secure web system independently administered by

a trials unit to ensure concealment of allocation. Allocation was minimised on OCD severity,

OCD duration, antidepressant use, and depression severity. It was not possible to mask partici-

pants or clinical staff to treatment allocation. Research staff undertaking assessments were

masked to allocation: unmasking was reported in 30%, 22%, and 26% of the 3, 6, and 12

month interviews, respectively. When this occurred, subsequent assessments were done by

another researcher to limit bias.

Procedures

cCBT was delivered using OCFighter (www.ccbt.co.uk), a commercial OCD program.

OCFighter involves 9 steps (focussed on exposure and response prevention) to help people

with OCD carry out treatment and monitor progress. Participants received a secure login and

were advised to use cCBT at least 6 times over 12 weeks. Participants received six 10-minute

telephone calls, for risk assessment, progress review, and problem solving.

Guided self-help was delivered using the book Obsessive Compulsive Disorder: A Self-HelpBook [18], which is focussed on exposure and response prevention. Participants received

weekly guidance, with an initial session (60 minutes face to face or by telephone, dependent on

patient preference) followed by 10 30-minute sessions over 12 weeks. The support involved an

explanation of the workbook, help devising goals, risk assessment, support for conducting

CBT homework, progress review, and problem solving.

Support for both cCBT and guided self-help was provided by “psychological well-being

practitioners.” These are graduates with no prior clinical qualifications who receive 12 months

training and who are responsible for delivering guided self-help CBT and general education

for anxiety and depression in England. Most have limited OCD-specific training. They were

trained in both interventions over 3 days by the research team (with additional support from

the company supplying cCBT). All staff received telephone supervision every fortnight from

the research team or from experienced therapists within routine services. In total, 93 psycho-

logical well-being practitioners managed patients in the trial (range 1–18 patients), with 46

practitioners allocated patients in both interventions. Psychological well-practitioner charac-

teristics are reported in Table A in S1 Appendix.

Psychological well-being practitioners recorded dates, length, and mode of contact for all

sessions and were asked to record sessions to examine fidelity. We also received automated

recordings of cCBT use. Fidelity was evaluated by an independent rater blind to outcome. We

defined tasks to be carried out in both interventions, which were coded from recordings as

“implicit,” “explicit,” or “absent,” and an overall rating was generated using a 5-point scale

(“unacceptable” to “excellent”).

The comparator was waiting list for therapist-led CBT. As the trial was a pragmatic design

within routine services, we were unable to mandate a waiting period for therapist-led CBT.

We expected that most patients would start therapist-led CBT 3–6 months following alloca-

tion, after receiving their low-intensity interventions (where allocated). Therapist-led CBT was

typically 8–20 face-to-face, 45–60-minute weekly sessions.



http://www.ccbt.co.uk/


In this pragmatic trial, we placed no restrictions on treatment after randomisation. Before

seeing a therapist, patients on waiting lists sometimes received interventions other than our

low-intensity interventions, including education, medication, or nonspecific interventions

(such as “stress management”). All additional care outside the trial protocol was recorded as

part of the economic evaluation.

Clinical outcomes

We conducted follow-up assessments at 3 months (primary outcome timepoint), 6 months,

and 12 months following randomisation. The primary outcome measure, Yale-Brown Ob-

sessive Compulsive Scale–Observer Rated (Y-BOCS-OR) [17], was collected face to face at

baseline. At follow-up time points where face-to-face collection was not possible following a

highly structured standardised operating procedure, telephone or postal assessment using the

Y-BOCS-SR was attempted.

The Y-BOCS-OR is an interview-administered structured assessment that provides an indi-

cation of OCD symptom severity. It consists of 2 comprehensive symptom checklists exploring

current and past symptoms of obsession and compulsion (over the past week and past symp-

toms) and a 10-item severity scale exploring current obsessive and compulsive symptoms.

Impairment over 5 clinical domains is identified: time consumed, functional impairment, psy-

chological distress, efforts to resist, and perceived sense of control on a 5-point Likert scale

from 0 (none) to 4 (extreme). Scores from the 10 items are summed to identify level of severity

(0–7 subclinical, 8–15 mild, 16–23 moderate, 24–31 severe, and 32–40 extreme).

Secondary outcomes were collected at baseline and at the 3, 6, and 12-month follow-up.

Outcomes included the Y-BOCS-SR, a self-report version of the Y-BOCS-OR scale. When it

was not possible to complete the Y-BOCS-OR, the Y-BOCS-SR was used as a proxy.

Other secondary outcomes (all self-report) included the Short Form-36 (SF-36) [19] for

health-related quality of life; Clinical Outcomes in Routine Evaluation (CORE-OM) [20] for

distress; Patient Health Questionnaire (PHQ-9) [21] for depression; Generalized Anxiety Dis-

order 7-item (GAD-7) scale [22] for generalised anxiety disorder; Work and Social Adjustment

Scale (WSAS) [23] for functional impairment; IAPT Employment Status Questions (A13-A14)

[24] for employment rates and receipt of statutory sick pay; and the Client Satisfaction Ques-

tionnaire (CSQ-8-UK) [25] for satisfaction. Comorbidities (Clinical Interview Schedule-

Revised; CIS-R) [26] and demographics were collected at baseline only.

Statistical analysis

With 3 pair-wise comparisons, alpha was set at 1.67%. We assumed a standard deviation for

the primary outcome of 7.3, a correlation between baseline and follow up of 0.43 [27], and a

therapist intracluster correlation (ICC) between therapists (0.06) and within therapist (0.015).

Assuming 85% retention, 432 patients were required. Trial monitoring suggested a lower fol-

low-up rate, and thus, the sample size was increased to 473 to retain power. In total, 366

patients at follow-up provided power greater than 80% to detect a clinically significant differ-

ence of 3 Y-BOCS points for each comparison.

Preliminary modelling determined methods for handling missing data (full details provided

in the statistical analysis plan: https://dx.doi.org/10.6084/m9.figshare.3503885). There was no

deviation from the prespecified plan, all analyses were conducted, and those not present in this

manuscript are reported in the Health Technology Assessment (HTA) report [28]. Missing

baseline covariates were imputed by single imputation [29] using other covariates. Analyses of

the primary outcome were based on a linear mixed model with random effects for psychologi-

cal well-being practitioners. As practitioners were crossed with treatment, correlated random



https://dx.doi.org/10.6084/m9.figshare.3503885


effects were included for each treatment, enabling estimation of the ICC for cCBT and guided

self-help. We included the following covariates: OCD duration and severity; anxiety, depres-

sion, antidepressant use; and gender. Binary outcomes (e.g., uptake of therapist-led CBT) used

logistic regression to estimate adjusted odds ratios (ORs), with the same baseline covariates.

Analysis used intention-to-treat subject to the availability of data. Distributional assumptions

of the models were checked. All outcomes included in the Obsessive Compulsive Treatment

Efficacy Trial (OCTET) protocol are detailed.

Patient involvement

Patients and members of the public were involved throughout the trial, including the design, man-

agement, and conduct of the trial. From the outset, the chief executive of a national user-led orga-

nisation (Anxiety UK) was involved as a coapplicant and collaborator. Members of an OCD self-

help group assisted with the development of the guided self-help manual and adaptations to one

of the trial outcome measures. A service user with OCD was a member of the trial steering com-

mittee, while another conducted some of the patient acceptability interviews. The findings have

been disseminated to trial participants, and the results presented at a national user conference.

Results

Recruitment, retention, and baseline characteristics

We opened recruitment in 15 clinical sites in England between February 2011 and May 2014,

with the last follow-up in May 2015. There were 2 postrandomisation exclusions: one aged

under 18 years and one at risk of suicide. In total, 473 eligible patients were randomised (see

Fig 1). Baseline sociodemographic characteristics are presented in Table 1, with baseline clini-

cal data presented in Table 2. Data were indicative of severe OCD, mild-to-moderate depres-

sion, and moderate anxiety. Just over half reported previous professional help with OCD, and

around half were using antidepressants. More than half reported OCD for more than 10 years.

There were no substantial differences at baseline.

Treatment fidelity and adherence

Patient flow is shown in Figs 1 and 2. Retention rates were 81% at 3 months, 75% at 6 months,

and 71% at 12 months and were broadly similar across arms (Fig 1). Contrary to expectation,

approximately 29% of patients started to access therapist-led CBT prior to the 3-month assess-

ment (Fig 2). More detailed data on CBT uptake and predictors of uptake are detailed sepa-

rately (Tables B and C in S1 Appendix).

Fifty-nine percent of participants had at least 1 session with a psychological well-being prac-

titioner in cCBT. The mean number of sessions was 2.3 (SD 2.5), and the average length was

13.4 minutes (93% by telephone). Of the 9 cCBT steps, the mean number completed was 3.7

(SD 3.2). Sixty-five percent of participants had at least 1 session with a psychological well-

being practitioner in guided self-help. The mean number of sessions was 4.1 (SD 4.3) over 57

minutes for session 1 and 31 minutes for sessions 2–11 (48% face-to-face, 26% telephone, 22%

both, 4% missing). Rates of recording for fidelity assessment were low (26% guided self-help,

17% cCBT). Of the sessions recorded in cCBT, 11 (65%) were rated “good” and 6 (35%) as

“excellent.” Of the sessions in guided self-help, 9 (21%) were rated “satisfactory,” 24 (56%)

rated “good,” and 10 (23%) “excellent.”

Table 3 gives the summary statistics for the primary (Y-BOCS-OR) and selected secondary

outcomes (Y-BOCS-SR, CSQ-8, and EuroQol five dimensions questionnaire [EQ-5D]). Com-

plete outcomes are reported separately (Table D in S1 Appendix).




Does access to guided self-help or cCBT provide more rapid

improvement in OCD symptoms at 3 months compared to a waiting list

for therapist-led CBT?

There was no significant benefit of access to cCBT (adjusted mean difference = −0.71, 95% CI

−2.12 to 0.70, p = 0.325). There was statistically significant benefit of guided self-help (adjusted

mean difference = −1.91, 95% CI −3.27 to −0.55, p = 0.006), although the effect was less than

the prespecified “clinically important difference” of 3 points.

Analyses of secondary outcomes (Tables E, F, and G in S1 Appendix) showed only 1 signifi-

cant difference, an effect of cCBT on anxiety (adjusted mean difference = −1.50, 95% CI −2.67

to −0.33, p = 0.012).

Does access to guided self-help or cCBT improve patient satisfaction at

3 months compared to a waiting list for therapist-led CBT?

Satisfaction data are shown in Table 3. There were no differences in patient satisfaction

among patients receiving cCBT compared to those allocated to a waiting list (adjusted mean

Fig 1. Consolidated Standards of Reporting Trials (CONSORT) flow chart illustrating recruitment. Abbreviations: cCBT, computerised

cognitive-behaviour therapy; m, month; QALY, quality-adjusted life year.

https://doi.org/10.1371/journal.pmed.1002337.g001





difference = −0.31, 95% CI −2.07 to 1.45, p = 0.732). Patients receiving guided self-help tended

to be more satisfied than those allocated to a waiting list for therapist-led CBT (adjusted mean

difference = 1.69, 95% CI −0.04 to 3.42, p = 0.055), although the estimate did not reach signifi-

cance according to the corrected significance level. Patients receiving cCBT were less satisfied

than those receiving guided self-help (adjusted mean difference = −2.00, 95% CI −3.63 to

−0.37, p = 0.016).

Table 1. Baseline sociodemographic variables.

Characteristic cCBT n = 157 Guided self-help n = 158 Waiting list n = 158

Age (years): median (range) 32.0 (18–77) 32.8 (18–72) 33.3 (19–66)

frequency (%) frequency (%) frequency (%)

Sex

Male: n (%) 66 (42.0%) 57 (36.1%) 65 (41.1%)

Female: n (%) 91 (58.0%) 101 (63.9%) 93 (58.9%)

Ethnicity

White: n (%) 145 (92.4%) 154 (97.5%) 149 (94.3%)

Nonwhite: n (%) 12 (7.6%) 4 (2.5%) 8 (5.1%)

Missing: n (%) 0 (0%) 0 (0%) 1 (0.6%)

Marital Status

Married/living with partner: n (%) 84 (53.5%) 81 (51.3%) 85 (53.8%)

Other: n (%) 70 (44.6%) 75 (47.4%) 73 (46.2%)

Missing: n (%) 3 (1.9%) 2 (1.3%) 0 (0%)

Employment status*

Employed: n (%) 86 (54.4%) 95 (60.1%) 97 (61.4%)

Unemployed and seeking work: n (%) 10 (6.3%) 14 (8.9%) 9 (5.7%)

Student: n (%) 17 (10.8%) 19 (12.0%) 17 (10.8)

Long-term sick/disabled receiving income support or incapacity benefit: n (%) 22 (14.0%) 23 (14.6%) 15 (9.5%)

Homemaker—not actively seeking work: n (%) 15 (9.6%) 9 (5.7%) 11 (7.0%)

Not receiving benefits and not actively seeking work: n (%) 1 (0.6%) 0 (0%) 1 (0.6%)

Unpaid voluntary work and not actively seeking work: n (%) 1 (0.6%) 1 (0.6%) 0 (0%)

Retired: n (%) 6 (3.8%) 5 (3.2%) 6 (3.8%)

Missing: n (%) 2 (1.3%) 1 (0.6%) 5 (3.2%)

Receiving statutory sick pay

Yes: n (%) 8 (5.1%) 8 (5.1%) 11 (7.0%)

No: n (%) 144 (91.7%) 146 (92.4%) 138 (87.3%)

Missing: n (%) 5 (3.2%) 4 (2.5%) 9 (5.7%)

Accessed previous OCD help

Yes: n (%) 76 (48.4%) 86 (54.4%) 72 (45.6%)

No: n (%) 80 (51.0%) 71 (44.9%) 85 (53.8%)

Missing: n (%) 1 (0.6%) 1 (0.7%) 1 (0.6%)

Education

Below degree level: n (%) 107 (68.2%) 110 (69.6%) 112 (70.9%)

Degree level or higher: n (%) 45 (28.6%) 43 (27.2%) 40 (25.3%)

Missing: n (%) 5 (3.2%) 5 (3.2%) 6 (3.8%)

* N and % for all groups do not sum (i.e., to sample size or 100%). This is as a result of some participants indicating more than 1 employment status.

Abbreviations: cCBT, computerised cognitive-behaviour therapy; OCD, obsessive-compulsive disorder.

https://doi.org/10.1371/journal.pmed.1002337.t001





Table 2. Baseline clinical characteristics.


frequency (%) frequency (%) frequency (%)

Current antidepressant medication 82 (52.2%) 81 (51.3%) 80 (50.6%)

OCD chronicity

0–5 years 53 (33.8%) 52 (32.9%) 51 (32.3%)

6–9 years 18 (11.5%) 18 (11.4%) 19 (12.0%)

�10 years 86 (54.8%) 88 (55.7%) 88 (55.7%)

Comorbidity (primary diagnosis)

Mixed anxiety and depressive disorder 23 (14.6%) 23 (14.6%) 15 (9.5%)

Mild depressive disorder 18 (11.5%) 18 (11.4%) 20 (12.7%)

Moderate depressive disorder 28 (17.8%) 24 (15.2%) 26 (16.5%)

Severe depressive disorder 7 (4.5%) 13 (8.2%) 12 (7.6%)

Generalised anxiety disorder 18 (11.5%) 27 (17.1%) 18 (11.4%)

Specific phobia 10 (6.4%) 6 (3.8%) 6 (3.8%)

Social phobia 2 (1.3%) 1 (1%) 0 (0%)

Agoraphobia 0 (0%) 2 (0.6%) 3 (1.9%)

Panic disorder 2 (1.3%) 0 (0%) 5 (3.2%)

Mean (SD) n Mean (SD) n Mean (SD) n

Y-BOCS-ORa 25.03 (5.45) 157 25.01 (5.02) 158 25.34 (5.44) 158

Median 25 25 25

Min; Max 13; 38 14; 39 13; 38

Missing 0 0 0

Y-BOCS-SR

Mean (SD) 24.34 (5.1) 24.18 (4.82) 24.20 (4.99)

Median 24 24 24

Min; Max 16;36 16;40 16;38

PHQ-9

Mean (SD) 11.90 (6.27) 155 11.40 (6.56) 154 11.93 (6.29) 154

Median 12 11.5 12

Min; Max 0; 27 0; 26 0; 26

GAD-7

Mean (SD) 12.90 (5.33) 12.72 (5.56) 12.52 (5.52)

Median 13 14 13

Min; Max 2; 21 1; 21 0; 21

Missing 2 4 4

CORE-OM

Mean (SD) 15.95 (6.27) 15.23 (6.67) 15.79 (6.63)

Median 16 16 16

Min; Max 5; 35 1; 34 1; 33

Missing 3 3 5

SF36–PCS

Mean (SD) 54.39 (11.29) 54.18 (9.57) 54.09 (10.56)

Median 57.36 56.01 57.14

Min; Max 18.04; 71.89 17.59; 70.35 22.21; 72.23

Missing 3 3 5

SF36–MCS

Mean (SD) 32.89 (9.87) 33.86 (11.05) 33.23 (11.71)

Median 32.66 34.33 33.17

(Continued)




Does access to guided self-help or cCBT prior to therapist-led CBT

provide longer-term improvement in OCD symptoms at 12 months

compared to therapist-led CBT alone?

There was no significant long-term benefit from access to either guided self-help or cCBT

(cCBT adjusted mean difference = −1.37, 95% CI −3.59 to 0.84, p = 0.224; guided self-help

adjusted mean difference −2.37, 95% CI −4.37 to −0.38, p = 0.02; Table 2).

Table 2. (Continued)


Min; Max 11.88; 59.52 7.30; 58.55 10.64; 65.08

Missing 3 3 5

WSAS

Mean (SD) 14.78 (9.85) 15.05 (10.54) 14.74 (9.66)

Median 13 14 13

Min; Max 2; 21 1; 21 0; 21

Missing 2 4 4

a incorporating Y-BOCS-SR if Y-BOCS-OR missing or incomplete. Patients were included in the trial with a Y-BOCS-SR of 16+, so all patients had a

Y-BOCS-SR of 16+ at baseline, but some had a Y-BOCS-OR of less than 16.

Abbreviations: cCBT, computerised cognitive-behaviour therapy; CORE-OM, Clinical Outcomes in Routine Evaluation; GAD-7, Generalised Anxiety

Disorder 7-item; MCS, Mental Component Summary; OCD, obsessive-compulsive disorder; PCS, Physical Component Summary; SF36, Short Form-36;

WSAS, Work and Social Adjustment Scale; Y-BOCS-OR, Yale-Brown Obsessive Compulsive Scale–Observer-Rated; Y-BOCS-SR, Yale-Brown Obsessive

Compulsive Scale–Self-Report.


Fig 2. Flow chart illustrating therapist-led CBT uptake. Abbreviations: cCBT, computerised cognitive-behaviour therapy; OCD, obsessive-

compulsive disorder.







Tab

le3.

Ou

tco

me

measu

resu

mm

ary

sta

tisti

cs

an

din

terv

en

tio

neff

ects

at3,6,an

d12

mo

nth

s.

Su

pp

ort

ed

cC

BT

Gu

ided

self

-help

Wait

ing

list

Su

pp

ort

ed

cC

BT

–W

ait

ing

list

Gu

ided

self

-help

–W

ait

ing

list

Su

pp

ort

ed

cC

BT

–G

uid

ed

self

-help

Mean

SD

nM

ean

SD

nM

ean

SD

nA

dju

ste

dm

ean

dif

fere

nce

+

95%

CI

pA

dju

ste

dm

ean

dif

fere

nce

+

95%

CI

pA

dju

ste

dm

ean

dif

fere

nce

+

95%

CI

p

Y-B

OC

S-O

R

Baselin

e25.0

35.4

5157

25.0

15.0

2158

25.3

45.4

4158

3m

onth

s21.1

66.8

9121

20.1

96.8

3130

22.1

86.5

4132

−0.7

1(−

2.1

2to

0.7

0)

0.3

25

−1.9

1(−

3.2

7to

−0.5

5)

0.0

06*

1.2

0(−

0.2

2to

2.6

1)

0.0

97

6m

onth

s18.9

67.2

6112

18.7

07.7

0122

20.2

97.2

7122

−1.1

3(−

2.8

4to

0.5

8)

0.1

95

−1.3

2(−

3.0

0to

0.3

5)

0.1

21

0.1

9(−

1.5

1to

1.9

0)

0.8

24

12

month

s16.1

48.6

9105

15.1

98.3

5113

17.9

38.0

7114

−1.3

7(−

3.5

9to

0.8

4)

0.2

24

−2.3

7(−

4.3

7to

−0.3

8)

0.0

20

1.0

0(−

1.1

9to

3.1

9)

0.3

71

Y-B

OC

S-S

R

Baselin

e24.3

45.1

0157

24.1

84.8

2158

24.2

04.9

9158

3m

onth

s20.4

67.0

6119

19.8

06.9

0128

20.8

86.4

8127

−0.4

3(−

1.7

9to

0.9

3)

0.5

31

−1.3

1(−

2.6

5to

0.0

4)

0.0

56

0.8

7(−

0.4

9to

2.2

3)

0.2

09

6m

onth

s18.6

07.4

7110

18.2

97.7

8119

19.3

47.2

4118

−0.8

7(−

2.5

2to

0.7

8)

0.3

00

−1.1

7(−

2.8

7to

0.5

3)

0.1

78

0.3

0(−

1.4

2to

2.0

2)

0.7

35

12

month

s15.6

18.7

0101

15.7

28.1

1109

17.3

88.2

4107

−1.4

5(−

3.6

7to

0.7

6)

0.1

98

−1.5

2(−

3.5

4to

0.4

9)

0.1

37

0.0

7(−

2.0

1to

2.1

6)

0.9

46

CS

Q-8

3m

onth

s22.4

15.9

192

24.3

75.4

9101

22.7

56.0

683

−0.3

1(−

2.0

7to

1.4

5)

0.7

32

1.6

9(−

0.0

4to

3.4

2)

0.0

55

−2.0

0(−

3.6

3to

−0.3

7)

0.0

16*

6m

onth

s23.7

55.7

881

24.2

66.2

982

24.6

45.8

077

−0.8

7(−

2.7

7to

1.0

3)

0.3

71

−0.3

1(−

2.2

3to

1.6

1)

0.7

51

−0.5

6(−

2.4

4to

1.3

2)

0.5

61

EQ

-5D

Baselin

e0.6

70.2

9155

0.6

80.2

6155

0.6

80.2

6154

3m

onth

s0.6

90.3

1104

0.7

30.2

4117

0.6

70.2

8124

0.0

3(−

0.0

2to

0.0

9)

0.2

46

0.0

5(0

.00

to

0.1

0)

0.0

42

−0.0

1(−

0.0

7to

0.0

4)

0.4

91

6m

onth

s0.7

10.2

994

0.7

20.2

5105

0.7

10.2

5106

0.0

0(−

0.0

6to

0.0

6)

0.9

61

0.0

0(−

0.0

5to

0.0

7)

0.8

15

−0.0

0(−

0.0

7to

0.0

6)

0.8

64

12

month

s0.7

90.2

784

0.7

30.2

6100

0.7

00.3

199

0.0

7(0

.00

to

0.1

5)

0.0

41

0.0

3(−

0.0

4to

0.0

9)

0.4

49

0.0

5(−

0.0

2to

0.1

2)

0.1

79

+M

ean

diffe

rence

adju

ste

dfo

rY

-BO

CS

-OR

,P

HQ

-9,G

AD

-7,antidepre

ssantuse,gender,

and

OC

Ddura

tion

(0–5,6–9,and�

10

years

).

*T

he

sig

nifi

cance

leveli

ssetat1.6

7%

toadju

stfo

r3

pair-w

ise

com

parisons.

Abbre

via

tions:C

SQ

-8,C

lientS

atisfa

ction

Questionnaire;E

Q-5

D,E

uro

Qolfive

dim

ensio

ns

questionnaire;G

AD

-7,G

enera

lised

Anxie

tyD

isord

er7-ite

m;O

CD

,obsessiv

e-c

om

puls

ive

dis

ord

er;

PH

Q-9

,P

atientH

ealth

Questionnaire;Y

-BO

CS

-OR

,Y

ale

-Bro

wn

Obsessiv

eC

om

puls

ive

Scale

–O

bserv

er-

Rate

d;Y

-BO

CS

-SR

,Y

ale

-Bro

wn

Obsessiv

eC

om

puls

ive

Scale

–

Self-R

eport

.

htt

ps:

//doi.o

rg/1

0.1

371/jo

urn

al.p

med

.1002337.t003





As a post hoc analysis, we tested whether low-intensity interventions were formally nonin-

ferior to waiting list for therapist-led CBT at 12 months. A 98.33% confidence interval corre-

sponds to a 1.67% significance level that we have used for hypothesis testing. For the

comparison of cCBT against waiting list, the 98.33% confidence interval is −4.07 to 1.33, and

for guided self-help against waiting list, it is −4.81 to 0.06. Given that the upper limits are sub-

stantially smaller than the prespecified criterion for a clinically important difference (3 points),

we conclude that both interventions are noninferior to waiting list at 12 months.

Does access to guided self-help or cCBT reduce uptake of therapist-led

CBT over 12 months?

Therapist-led CBT uptake is shown in Fig 2. Both interventions were associated with signifi-

cantly lower uptake of therapist-led CBT at 12 months (cCBT adjusted OR = 0.34, 95% CI 0.15

to 0.79 p = 0.011; guided self-help adjusted OR = 0.27, 95% CI 0.12 to 0.60 p = 0.001) (Table 4).

Post hoc, we compared intervention use and 12-month OCD outcomes among guided self-

help and cCBT patients who did and did not access therapist-led CBT (Table H in S1

Table 4. Logistic regression model for CBT uptake at 6 and 12 months.

Comparison Adjusted odds ratio 95% CI p

6 months

Group cCBT versus WL 0.42 (0.24–0.73) 0.002*

GSH versus WL 0.48 (0.22–1.03) 0.06

cCBT versus GSH 0.87 (0.42–1.84) 0.718

Baseline outcome measures Y-BOCS-OR 1.02 (0.97–1.06) 0.514

GAD-7 0.99 (0.94–1.04) 0.591

PHQ-9 1.02 (0.98–1.07) 0.271

Antidepressant medication Yes 0.71 (0.46–1.09) 0.117

Duration of OCD 6–9 years 1.26 (0.60–2.64) 0.552

10 or more years 0.89 (0.55–1.42) 0.619

Gender Male 1.12 (0.73–1.73) 0.606

Exponential Function (Constant) 2.14 (0.67–6.82) 0.201

12 months

Group cCBT versus WL 0.34 (0.15–0.79) 0.011*

GSH versus WL 0.27 (0.12–0.60) 0.001*

cCBT versus GSH 1.27 (0.53–3.00) 0.59

Baseline outcome measures Y-BOCS-OR 1.03 (0.97–1.08) 0.36

GAD-7 1.03 (0.97–1.08) 0.341

PHQ-9 0.99 (0.94–1.04) 0.73

Antidepressant medication Yes 1.02 (0.63–1.67) 0.933

Duration of OCD 6–9 years 2.66 (1.03–6.89) 0.043

10 or more years 0.99 (0.59–1.67) 0.968

Gender Male 1.25 (0.76–2.03) 0.395

Exponential Function (Constant) 2.86 (0.76–10.81) 0.121

* Note, the Bonferroni corrected significance level is 1.67%, for 3 pair-wise comparisons.

** Note, results are taken from a logistic regression model and any “effect” should be interpreted as an odds

ratio.

Abbreviations: cCBT, computerised cognitive-behaviour therapy; GAD-7, Generalised Anxiety Disorder

7-item; GSH, guided self-help; OCD, obsessive-compulsive disorder; PHQ-9, Patient Health Questionnaire;

Y-BOCS-OR, Yale-Brown Obsessive Compulsive Scale–Observer-Rated; WL, waiting list.






Appendix). Although lacking randomisation, the data do not suggest that those who accessed

only guided self-help or cCBT demonstrated markedly worse outcomes than those who

accessed both a low-intensity intervention and therapist-led CBT (Table I in S1 Appendix).

Discussion

Principal outcomes

We assessed the role of 2 low-intensity interventions (guided self-help and cCBT) in OCD.

Prior to access to therapist-led CBT, guided self-help demonstrated statistically significant

benefits over the waiting list, but the difference did not meet the prespecified criterion for clin-

ical significance. cCBT did not demonstrate significant benefit at the 3- or 12-month follow-

up. Access to low-intensity interventions does not provide more rapid symptom relief.

Over 12 months, access to low-intensity interventions prior to therapist-led CBT did not

significantly augment the effects of therapist-led CBT on OCD symptoms in the longer term.

Rapid access to low-intensity interventions did lead to significant reductions in uptake of ther-

apist-led CBT, which did not compromise patient outcomes at 12 months.

Strengths and limitations

To our knowledge, we conducted the largest trial of psychological therapy for OCD worldwide.

We achieved acceptable levels of retention. When patients were not able to provide the pri-

mary clinical outcome using the observer-reported version, we used self-report as a proxy.

These different measures show high associations [30,31], with some evidence of lower scores

in the self-reported version, but proxy measures were only used in 8% and 11% of cases at 3

and 12 months, with minimal differences in rates of use between arms. In this pragmatic trial,

recruitment was over multiple sites and involved a large number of psychological well-being

practitioners. This enhances external validity, as delivery was not restricted to a small number

of specialised sites or highly selected professionals. However, many psychological well-being

practitioners only saw a few patients, which restricted the opportunity to practice their skills.

Uptake of the interventions was reasonable (65% guided self-help and 59% cCBT). Collecting

detailed data on fidelity proved difficult, but analysis of the data provided some evidence that

delivery of guided self-help and cCBT was in line with protocols.

Several issues are worth noting in this pragmatic design. First, we did not mandate a

defined waiting time for therapist-led CBT, although the expectation was 3–6 months. In prac-

tice, around 40% of patients allocated to a waiting list for therapist-led CBT started to receive

some contact with their therapist before 3 months, compared with just over 20% in the guided

self-help and cCBT groups. This would reduce differences in outcomes between guided self-

help, cCBT, and the waiting list comparator at 3 months, leading to conservative estimates of

effect. Still, these data refer to patients receiving any contact with therapist-led CBT, which in

most cases would involve an initial session or two, rather than a full “dose” of treatment. Nev-

ertheless, relatively quick access to CBT in the waiting list arm would have reduced short-term

benefit associated with the low-intensity interventions. The effects of low-intensity interven-

tions may have been more pronounced at 3 months if CBT was less accessible than in the cur-

rent trial. The longer-term analyses are less affected, as all patients were expected to receive

both a low-intensity intervention (where allocated) and therapist-led CBT over 12 months.

We have shown that uptake of therapist-led CBT was lower in the groups allocated to low-

intensity treatment. This could reflect positive outcomes from some aspects of the low-inten-

sity interventions, although our analysis showed that this was largely restricted to patient

satisfaction rather than clinical benefits. Even in the absence of significant clinical benefits,

providing low-intensity treatments may give patients a sense of support and progress. When




combined with natural improvement in symptoms over time (as found in all groups), this may

mean that patients do not feel a need for further intensive support or no longer wish to engage

with services. However, the numbers of patients attending therapist-led CBT increased in all

groups over time. It is possible that, had our follow-up been longer than 12 months, eventual

uptake of therapist-led CBT across all groups would be the same.

Secondly, we placed no restrictions on medication use, and in line with most psychological

therapy trials in OCD, a proportion of patients were taking medication. Baseline self-reported

use of antidepressant treatment is provided in Table 1, showing that about half the patients

reported using antidepressants, with no differences between groups. Data on antidepressant

use after allocation showed that, over the 12-month period of the trial, self-reported use of

antidepressants decreased (cCBT 26%, guided self-help [GSH] 32%, and waiting list 27%) [28].

We do not have details of the nature or quality of antidepressant prescription. Although anti-

depressants are effective in OCD [3,32], it seems unlikely that such small differences between

arms would be a major driver of study outcomes.

Thirdly, there was no attempt to match the level or type of clinician contact across the 2

low-intensity interventions: indeed, the study was specifically designed to test the relative

value of 2 different forms of low-intensity intervention. GSH involved both a different delivery

format and more clinician contact, so our trial is not a strict comparison of paper and digital

interventions. Although increased clinician contact may well enhance acceptability and effec-

tiveness, its additional costs were accounted for in the economic analysis.

Fourth, we did not undertake quality assurance of the therapist-led CBT provided to all

patients. As noted above, therapist-led CBT was provided by a range of practitioners in a range

of areas and is likely to be reasonably representative of the treatment provided in the National

Health Service (NHS) in England, which remains optimal for a pragmatic trial. Formal mea-

surement of quality would have been preferable but logistically complex.

The trial adopted aspects of a stepped care model, whereby patients are offered a low-inten-

sity intervention first, with a proportion progressing to therapist-led CBT. However, unlike

true stepped care, there was no regular assessment of outcome, and access to therapist-led

CBT was available to all, rather than as part of a defined “stepping” mechanism following non-

response to treatment or deterioration. Therefore, our analysis does not assess the benefits of

low-intensity treatment in a full stepped care model.

Interpretation of the results in the context of the wider literature

At the initiation of this trial, the evidence base was very limited [11]. While the current trial

was being delivered, a number of additional studies were published. One small trial (n = 56)

used similar interventions to the present study (GSH and supported cCBT) and compared

their effects in a group of volunteers recruited through a website. Very large effects were

found post-treatment [33]. A second trial (n = 86) exploring a minimally supported cCBT

intervention again found very large effects in a sample recruited online through a research cen-

tre [34]. A third trial randomised 34 patients with OCD to a supported internet-based writing

therapy and again found very large effects among a sample recruited via public notices. A long-

term follow-up [35] of a previous trial [12] showed enduring effects for cCBT and that “booster”

treatments were effective in maintaining gains. Finally, one trial (n = 128) explored the effects of

unsupported written material about metacognitive therapy in patients with OCD from internet

groups, self-help organisations, and clinical facilities and found small benefits [36].

The picture from these trials is more positive about the clinical benefits of “low-intensity”

treatments, especially supported cCBT, with most effect sizes over 0.5 and some exceeding 1.0.

This contrasts markedly with the modest clinical impacts observed in the current study. There




are a number of reasons that could account for these differences. The interventions do vary,

although it is not clear that the variation is large enough to account for the large variation in

effects. The current sample of patients have more severe symptoms at baseline (Y-BOCS scores

of 25, compared to 20–21 in the other studies), although again it is not clear that such modest

differences would be expected to lead to such profound variation in impact. Our current study

is far larger than the other trials, and some report quite large differences at baseline, which

can occur when small numbers of patients are randomised [33,36]. A potentially important

issue is the method of recruitment. The vast majority of the patients in the current study were

recruited through routine clinical services, whereas a number of the other trials used recruit-

ment through the internet; this may produce a sample with different clinical features and one

that is much more amenable to online cCBT interventions. Similar differences in effects

between large pragmatic trials in routine services and smaller trials recruiting through the

internet have been recently reported in depression [37].

Implications for service delivery

The trial demonstrated that neither form of low-intensity CBT was responsible for clinically

significant improvements in OCD symptoms among patients on the waiting list.

In the absence of any significant clinical benefit over waiting list only, readers may have

concerns about reductions in the use of therapist-led CBT at 12 months, as this might reflect

the substitution, or delay, of an evidence-based treatment. It may be that access to GSH or

cCBT means that patients are put off from engaging in subsequent therapist-led CBT. We

found no evidence that lower uptake of therapist-led CBT was associated with worse outcome

over 12 months. Concerns that GSH or cCBT inappropriately discourages patients from

engaging in subsequent therapist-led CBT are not supported by the wider literature, which

shows an increased likelihood of help seeking and greater healthcare use following self-help

[38,39]. It is possible that some patients who are offered GSH or cCBT improve so that they do

not need subsequent therapist-led CBT or make an informed choice to discontinue therapy

sooner rather than later. However, as noted earlier, differences in uptake between arms may

have reduced if a longer follow-up had been possible.

Our results raise questions about the targeting of low-intensity interventions. Recruiting

from waiting lists identified a sample with severe symptoms and a relatively long history of

treatment. Although most showed significant improvements over time (around 8–9 points on

the Y-BOCS across groups), the means at 12 months still showed significant symptoms

(around 16 points), meaning many would continue to be eligible for the trial. Routine provi-

sion of CBT within the NHS in line with clinical guidelines clearly leaves many patients with

clinically significant residual symptoms. Low-intensity treatments may be better targeted at a

less severely ill group, closer to the onset of their OCD. However, as this patient group is char-

acterised by late presentation to services, the viability of this is unclear.

Neither cCBT nor GSH showed clinically significant benefits at 3 months. Further develop-

ment of more effective low-intensity interventions may be required. Uptake of the interven-

tions was relatively low, although not abnormally so for a pragmatic trial. Both interventions

may benefit from enhancements that might improve motivation or adherence, which might

translate to greater clinical benefit.

The clinical results alone do not support an important role for low-intensity interventions

in the care pathway for OCD. However, full interpretation of the benefits of low-intensity

interventions for OCD also demands consideration of cost-effectivness, using comprehensive

assessments of costs, as well as appropriate measures of the impact of these interventions on




health-related quality of life and associated utility. We report the results of this analysis sepa-

rately [28].

Supporting information

S1 Appendix. Supplementary information file.

(DOCX)

S1 Text. HTA Obsessive Compulsive Treatment Efficacy Trial (OCTET) study protocol

v11 02.07.2015.

(DOC)

S2 Text. Consolidated Standards of Reporting Trials (CONSORT) checklist.

(DOC)

Acknowledgments

The authors would like to thank all the patients, health professionals, NHS Trust staff, and

Mental Health Research Network (MHRN) members who took part, provide support, and

contributed to this piece of research. We are very grateful to the members of our Trial Steering

Committee Data Monitoring Committee for their invaluable advice and support during the

project. We also acknowledge the vital contributions of study researchers for their assistance

with the conduct and analysis of the Psychological Wellbeing Practitioner (PWP) acceptability

interviews and fidelity assessment and acknowledge the statistician who drafted the statistical

analysis plan and set up the blinded interim analysis requested by the HTA.

Author Contributions

Conceptualization: KL PBo SB PBe DM SG LG GH SR MB PM NL CR.

Data curation: CR SB CF MH JG CA.

Formal analysis: CR SB CF MH.

Funding acquisition: KL PBo SB PBe DM SG LG GH SR MB PM NL CR.

Investigation: RP JM EP JKH OP JC FP.

Methodology: KL PBo SB PBe DM SG LG GH SR MB PM NL CR.

Project administration: JG CA.

Resources: JG SB CA MH CF CR.

Software: SB MH CF CR.

Supervision: KL JG PBe DM CA LG GH MB.

Validation: KL PBo JG SB PBe DM CA SG LG GH SR MB PM NL RP JM EP JKH OP JC MH

CF FP CR.

Visualization: KL PBo JG SB PBe DM CA SG LG GH SR MB PM NL RP JM EP JKH OP JC

MH CF FP CR.

Writing – original draft: KL PBo JG SB MH CR.

Writing – review & editing: KL PBo JG SB PBe DM CA SG LG GH SR MB PM NL RP JM EP

JKH OP JC MH CF FP CR.



http://journals.plos.org/plosmedicine/article/asset?unique&id=info:doi/10.1371/journal.pmed.1002337.s001




References1. National Institute for Health and Clinical Excellence. Obsessive-compulsive disorder: core interventions

in the treatment of obsessive-compulsive disorder and body dysmorphic disorder (Clinical guideline

31). London: NICE; 2005.

2. Hollander E, Broatch J, Himelein C, Rowland C, Stein D, Kwon J. Psychosocial function and economic

costs of obsessive-compulsive disorder. CNS Spect 1997; 2:16–25.

3. Skapinakis P, Caldwell D, Hollingworth W, Bryden P, Finberg N, Salkolvskis P, Welton N, Baxter H,

Kessler D, Churchill R and Lewis G. A systematic review of the clinical effectiveness and cost-effective-

ness of pharmacological and psychological interventions for the management of obsessive-compulsive

disorder in children/adolescents and adults. Health Technol Assess 2016; 20 (43). https://doi.org/10.

3310/hta20430 PMID: 27306503

4. Richards DA, Ekers D, McMillan D, Taylor RS, Byford S, Warren FC, et al. Cost and Outcome of Beha-

vioural Activation versus Cognitive Behavioural Therapy for Depression (COBRA): a randomised, con-

trolled, non-inferiority trial. Lancet Published Online July 22, 2016; http://dx.doi.org/10.1016/S0140-

6736(16)31140-0.

5. Andrews G, Cuijpers P, Craske M, McEvoy P, Titov N. Computer therapy for the anxiety and depressive

disorders is effective, acceptable and practical health care: a meta-analysis. PLoS ONE 2010;5(10):

e13196. https://doi.org/10.1371/journal.pone.0013196 PMID: 20967242

6. Cuijpers P, Donker T, Van Straten A, Li J, Andersson G. Is guided self-help as effective as face to face

psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of com-

parative outcome studies. Psychol Med 2010; 40:1943–57. https://doi.org/10.1017/

S0033291710000772 PMID: 20406528

7. Bower P, Gilbody S. Stepped care in psychological therapies: access, effectiveness and efficiency: nar-

rative literature review. Brit J Psychiat 2005; 186:11–17. https://doi.org/10.1192/bjp.186.1.11 PMID:

15630118

8. Lovell K, Ekers D, Fullford J, Baguley C, Bradshaw T. A pilot study of a self-help manual with minimal

therapist contact in the treatment of obsessive-compulsive disorder. Clin Eff Nurs 2005; 8(2):122–127.

9. Tolin DF, Diefenbach GJ, Maltby N, Hannan SE. Stepped care for obsessive-compulsive disorder: A

pilot study. Cogn Behav Pract 2005; 12(4):403–414.

10. Tolin KL, Hannan DF, Maltby S, Diefenbach N, Worhunsky P, Brady RE. A RCT of self-directed versus

therapist-directed CBT for OCD patients with prior medication trials. Behav Ther 2007; 38:179–191.

https://doi.org/10.1016/j.beth.2006.07.001 PMID: 17499084

11. Kaltenthaler E, Brazier J, De Nigris E, Tumur I, Ferriter M, Beverley C, et al. Computerised cognitive

behaviour therapy for depression and anxiety update: a systematic review and economic evaluation.

Health Technol Assess 2006; 10:1–168.

12. Andersson E, Enander J, Andren P, Hedman E, Ljotsson B, Bergstrom J, et al. Internet-based cognitive

behaviour therapy for obsessive-compulsive disorder: a randomized controlled trial. Psychol Med

2012;42(10):2193–203. https://doi.org/10.1017/S0033291712000244 PMID: 22348650

13. Loudon K, Treweek S, Sullivan F, Donnan P, Thorpe KE, Zwarenstein M. The PRECIS-2 tool: designing

trials that are fit for purpose. Brit Med J 2015; 350: h2147. https://doi.org/10.1136/bmj.h2147 PMID:

25956159

14. Sox HC, Lewis RJ. Pragmatic Trials: Practical Answers to “Real World” Questions. JAMA 2016; 316

(11):1205–1206. https://doi.org/10.1001/jama.2016.11409 PMID: 27654606

15. Gellatly J, Bower P, McMillan D, Roberts C, Byford S, Bee P, et al. Obsessive Compulsive Treatment

Efficacy Trial (OCTET) comparing the clinical and cost effectiveness of self-managed therapies: study

protocol for a randomised controlled trial. Trials 2014; 15(1):278.

16. Sheehan DV, Lecrubier Y, Harnett-Sheehan K, Amorim P, Janavs J, Weiller E et al. The Mini-Interna-

tional Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic

psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59 Suppl 20: 22–33.

17. Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL, et al. The Yale-Brown

obsessive compulsive scale: 1. Development, use and reliability. Arch Gen Psychiatry1989; 46:1006–

1011. PMID: 2684084

18. Lovell K, Gega L. Obsessive Compulsive Disorder: A Self-Help Book. The University of Manchester;

2011.

19. Ware JE, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): I. Conceptual frame-

work and item selection. Med Care 1992; 30:473–483. PMID: 1593914

20. Evans C, Connell J, Barkham M, Margison F, McGrath G, Mellor-Clark J, et al. Towards a standardised

brief outcome measure: psychometric properties and utility of the CORE-OM. Br J Psychiatry 2002; 180

(1):51–60.



https://doi.org/10.3310/hta20430

https://doi.org/10.3310/hta20430

http://www.ncbi.nlm.nih.gov/pubmed/27306503

http://dx.doi.org/10.1016/S0140-6736

http://dx.doi.org/10.1016/S0140-6736

https://doi.org/10.1371/journal.pone.0013196


https://doi.org/10.1017/S0033291710000772

https://doi.org/10.1017/S0033291710000772


https://doi.org/10.1192/bjp.186.1.11


https://doi.org/10.1016/j.beth.2006.07.001


https://doi.org/10.1017/S0033291712000244


https://doi.org/10.1136/bmj.h2147


https://doi.org/10.1001/jama.2016.11409





21. Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: the

PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire.

JAMA 1999; 282:1737–44. PMID: 10568646

22. Spitzer RL, Kroenke K, Williams JB, Lowe B. A brief measure for assessing generalized anxiety disor-

der: the GAD-7. Arch Intern Med 2006; 166:1092–1097. https://doi.org/10.1001/archinte.166.10.1092

PMID: 16717171

23. Mundt JC, Marks IM, Greist JH, Shear K. The work and social adjustment scale: a simple accurate mea-

sure of impairment in functioning. Br J Psychiatry 2002; 180:461–4. PMID: 11983645

24. The IAPT Data Handbook. Guidance on recording and monitoring outcomes to support local evidence-

based practice. 2015. URL: www.iapt.nhs.uk/silo/files/the-iapt-data-handbook.pdf (accessed 3 Novem-

ber 2011).

25. Larsen DL, Attkisson CC, Hargreaves WA, Nguyen TD. Assessment of client/patient satisfaction: devel-

opment of a general scale. Eval Program Plann 1979; 2(3): 197–207. PMID: 10245370

26. Lewis G, Pelosi AJ, Araya R, Dunn G. Measuring psychiatric disorders in the community: a standard-

ised assessment for use by lay interviewers. Psychol Med 1992; 22:465–486. PMID: 1615114

27. Lovell K, Cox D, Haddock G, Jones C, Raines D, Garvey R. Telephone administered cognitive behav-

iour therapy for treatment of obsessive compulsive disorder: Randomised controlled non-inferiority trial.

Brit Med J 2006; 333(883).

28. Lovell K, Bower P, Gellatly J, Byford S, Bee P, McMillan D, et al. Clinical, cost-effectiveness and accept-

ability of low intensity interventions in the management of OCD: the OCTET randomised controlled trial.

Health Technology Assessment [Online]. Available at: https://www.journalslibrary.nihr.ac.uk/

programmes/hta/098101/#/

29. White RI, Thompson SG. Adjusting for partially missing baseline measurements in randomized trials.

Stat Med 2005; 24(7): 993–1007. https://doi.org/10.1002/sim.1981 PMID: 15570623

30. Federici A, Summerfeldt LJ, Harrington JL, McCabe RE, Purdon CL, Rowa K, et al. Consistency

between self-report and clinician-administered versions of the Yale-Brown Obsessive-Compulsive

Scale. J Anxiety Disord 2010; 24(7): 729–33. https://doi.org/10.1016/j.janxdis.2010.05.005 PMID:

20561767

31. Steketee G., Frost R., & Bogart K. The Yale-Brown obsessive compulsive scale: interview versus self-

report. Behav Res Ther 1996; 34: 675–684. PMID: 8870295

32. Soomro GM, Altman D, Rajagopal S, Oakley-Browne M. Selective serotonin re-uptake inhibitors

(SSRIs) versus placebo for obsessive compulsive disorder (OCD). Cochrane Database of Systematic

Reviews 2008; Issue 1, CD00176.

33. Wootton BM, Dear BF, Johnston L, Terides MD, Titov N. Remote treatment of obsessive-compulsive

disorder: A randomized controlled trial. J Obsess-Compul Rel Disord 2013, 2(4), 375–384. http://doi.

org/10.1016/j.jocrd.2013.07.002

34. Mahoney AEJ, Mackenzie A, Williams AD, Smith J, Andrews G. Internet cognitive behavioural treat-

ment for obsessive compulsive disorder: A randomised controlled trial. Behav Res Ther 2014; 63:99–

106 https://doi.org/10.1016/j.brat.2014.09.012 PMID: 25461784

35. Andersson E, Steneby S, Karlsson K, Ljotsson B, Hedman E, Enander, et al. Long-term efficacy of

Internet-based cognitive behavior therapy for obsessive–compulsive disorder with or without booster: a

randomized controlled trial. Psychol Med 2014; 44(13):2877–87. https://doi.org/10.1017/

S0033291714000543 PMID: 25066102

36. Hauschildt M, Schroder J, Moritz S. Randomized-controlled trial on a novel (meta-) cognitive self-help

approach for obsessive-compulsive disorder (“myMCT”). J Obsess-Compul Rel Disord 2016;10:26–34.

37. Gilbody S, Littlewood E, Hewtii C, Brierley G, Tharmanathan P, Araya R, et al. Computerised cognitive

behaviour therapy (cCBT) as treatment for depression in primary care (REEACT trial): large scale prag-

matic randomised controlled trial Brit Med J 2015; 351:h5627 https://doi.org/10.1136/bmj.h5627 PMID:

26559241

38. Christensen H, Griffiths KM, Mackinnon AJ, Brittliffe K. Online randomized controlled trial of brief and

full cognitive behavior therapy for depression. Psychol Med 2006; 36, 1737e1746.

39. Wantland DJ, Portillo CJ, Holzemer WL, Slaughter R, McGhee EM. The Effectiveness of Web-Based

vs. Non-Web-Based Interventions: A Meta-Analysis of Behavioral Change Outcomes. J Med Internet

Res 2004; 6(4):e40. https://doi.org/10.2196/jmir.6.4.e40 PMID: 15631964




https://doi.org/10.1001/archinte.166.10.1092



http://www.iapt.nhs.uk/silo/files/the-iapt-data-handbook.pdf





https://doi.org/10.1002/sim.1981


https://doi.org/10.1016/j.janxdis.2010.05.005



http://doi.org/10.1016/j.jocrd.2013.07.002

http://doi.org/10.1016/j.jocrd.2013.07.002

https://doi.org/10.1016/j.brat.2014.09.012


https://doi.org/10.1017/S0033291714000543

https://doi.org/10.1017/S0033291714000543


https://doi.org/10.1136/bmj.h5627


https://doi.org/10.2196/jmir.6.4.e40



Documents

Lowintensity cognitivebehaviour therapy clinical ...centaur.reading.ac.uk/71064/6/journal.pmed.1002337.pdf · part of the peer-review process for funding, peer-reviewers identified